CN106612439A - Adaptive fast fractal image compression method - Google Patents
Adaptive fast fractal image compression method Download PDFInfo
- Publication number
- CN106612439A CN106612439A CN201610079621.2A CN201610079621A CN106612439A CN 106612439 A CN106612439 A CN 106612439A CN 201610079621 A CN201610079621 A CN 201610079621A CN 106612439 A CN106612439 A CN 106612439A
- Authority
- CN
- China
- Prior art keywords
- block
- image
- adaptive
- compression
- quantization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/99—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals involving fractal coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/625—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using discrete cosine transform [DCT]
Abstract
The invention proposes an adaptive fast fractal image compression method, relates to the technical field of computer information, and specifically relates to the field of image compression. The compression method can both perform adaptive compression on static images and moving images. The adaptive fast fractal image compression method combines an adaptive quantization method with a fast fractal theory, and comprises the following steps: firstly, judging a smooth region and a non-smooth region, and dividing an image into a plurality of sub-image blocks; secondly, adopting a JPEC algorithm to compress the sub-image blocks, performing DCT transform on the sub-image blocks and performing DPCM coding processing on a DC component to obtain an adaptive block; thirdly, building a domain block library according to the generated adaptive block; fourthly, segmenting a range block according to the built domain block library, extracting two features, namely a mean value and a variance of the adaptive block, using a cluster center of K range block classes to cluster, and obtaining corresponding K domain block classes; and finally, matching the K range block classes with the corresponding K domain block classes, and obtaining a fast fractal image compressed image of an adaptive image based on K-means clustering.
Description
Art
The present invention relates to computer information technology field, more particularly to compression of images field.
Background technology
Compression of images is the classical and popular research direction of image processing field.The purpose of compression of images is exactly on the premise of image effective information is not damaged, to remove the redundant data in image.At present in conventional image compression algorithm, an outstanding feature is exactly that image is uniformly split, and then carries out a series of conversion and coded treatment to each block, realizes the compression of image.Existing some Image Compression, however it remains that picture compression efficiency is not high, the problem that compression quality is bad.
Fractal theory is the extension of euclidean geometry correlation theory, and it describes the self-similarity of object in nature, what this self-similarity can be to determine, it is also possible to be in statistical significance.Fast Fractal Image compression is a kind of method for compressing image grown up on the basis of Fractal Geometry Theory, and its theoretical basis is iterated function system.Fast Fractal Image compress technique is set up present in natural image on the basis of local self-similarity, with the parameter of a compressed transform come phenogram picture.This compressed transform is made up of one group of mapping for acting on image subblock, discloses the local self-similarity of image presence.Due to storing the bit number of the bit number well below storage original image of affine transformation quantization parameter, it is possible to realizing the high power compression of view data.Fast fractal decoding is acted on any initial pictures to generate using novel Fast Iterative Procedure, reconstructed image by compressed transform iteration.It is several fracton figures by artwork predecomposition, so that subgraph has certain fractal structure, there is certain self affine feature between the entirety of subgraph and local, by substantial amounts of these subgraphs a point shape storehouse is constituted, each subgraph can find their Matching sub-image coding in these point of shape storehouse.So, image segmentation can be converted into the Fast Fractal algorithm of image, the coding of Matching sub-image is found in storehouse, finally throw away artwork, preserve subgraph coding, be stored or transmitted.
The content of the invention
For above-mentioned weak point, the present invention proposes a Novel Fast Fractal Image Compression Approach based on adaptive block, the theoretical basiss of the self adaptation splits' positions are that to be distributed in smooth area different with the characteristic of non-smooth area due to the frequency domain energy of image, the significant coefficient distribution for being mainly manifested in smooth area in discrete cosine transform domain is concentrated, and the non-smooth area of insignificant insignificant coefficient ratio is more.Although the significant coefficient distribution in non-smooth area is concentrated not as smooth area, but the adaptive quantizing method quantified to significantly value coefficient small step length with the big step size quantization of small size value coefficient can be set up by sequence, further entropy code is carried out according to itself statistical law to the significant coefficient and its address date after quantization, the compression ratio to non-smooth area can be improved, and Quality of recovery is also improved.
The purpose of the present invention is：Lift the compression efficiency of image and the reconstruction quality of compression image.
The present invention is adopted the technical scheme that for achieving the above object：A kind of self adaptation Novel Fast Fractal Image Compression Approach, realizes that process is as follows：
Step 1：Each label frequency of occurrences in sub-block is counted according to original image labelling matrix, maximum is obtained.Maximum is compared with decision threshold, more than decision threshold, then subimage block is judged to smooth area, otherwise continue to judge whether subimage block size is standard size, it is to divide the image into multiple subimage blocks according to quaternary tree mode, subimage block is scanned one by one, obtains the current subimage block after scanning.
Step 2：Current subimage block is processed using jpeg algorithm compression.After to the dimension dct transform of current subimage block 2, DC components carry out DPCM coded treatments, to sorting from big to small according to amplitude after AC component swepts, then quantify, encode, and generate adaptive block.Simultaneously to the closely related coding of address date, image adaptive process is realized.
Step 3：Adaptive block according to generating sets up domain blocks storehouse.The step-size in search of domain blocks is r, and removes the block of flat domain blocks, i.e. gray standard deviation less than certain given threshold value, and the completely available gray value of flat block replaces equal to the constant block of range block average, this reduces clusters number and search space.
Step 4：According to the domain blocks storehouse segmentation range block set up, gray standard deviation is considered as into flat block less than the range block of certain threshold value n, the constant block of range block average is directly equal to gray value to approach.Two features of average and variance of range block are extracted, using the two features K mean cluster is carried out, obtain K range block class.
Step 5：Two features of average and variance of adaptive block are extracted, adaptive block is clustered with the cluster centre of K range block class, obtain corresponding K domain blocks class.Finally K range block class and corresponding K domain blocks class are matched, fast typing compression image of the adapting to image based on K mean cluster is obtained.
The invention has the beneficial effects as follows：Lift the efficiency of compression of images and the quality of reconstruction image.
Description of the drawings
Accompanying drawing is the flow chart of self adaptation Novel Fast Fractal Image Compression Approach.
Specific embodiment
The present invention mainly describes a kind of NEW ADAPTIVE Novel Fast Fractal Image Compression Approach, including the self-adapting compressing to static and moving image.It is the innovation adaptive quantizing method on the basis of self adaptation splits' positions to the groundwork of still image self-adapting compressing research, and by itself and fast typing theory and combining, it is proposed that the NEW ADAPTIVE Novel Fast Fractal Image Compression Approach of the present invention.
Hereinafter, with reference to accompanying drawing, the present invention is described in detail.
First, self adaptation still image piecemeal, the flow chart of combining adaptive Novel Fast Fractal Image Compression Approach
In the Static Picture Compression algorithm based on self adaptation piecemeal, the core that smooth area and non-smooth area are self adaptation block algorithms is distinguished.Self adaptation method of partition based on edge information measure and based on potential function clustering fitted figure as histogrammic self adaptation method of partition is the core algorithm in self adaptation still image piecemeal.
In the present invention, two kinds of self adaptation still image block algorithms, to the subimage block of the various sizes of gained after piecemeal the quantization method quantification treatment of algorithm is adopted, and to the bulk smooth area obtained by piecemeal and the non-smooth area of fritter same quantization method is adopted.For smooth area, the quantization method of algorithm is very effective.This is because the table that quantifies that algorithm is recommended adopts little quantization step to low frequency part, big quantization step is adopted to medium-high frequency, the quantizing process of algorithm is the low-frequency component for retaining image, suppress the process of high frequency components, this is identical with the frequency domain energy characteristic distributions of smooth area, energy is concentrated mainly on low frequency part in smooth area, and high frequency components occupy back burner.So in the step of producing image adaptive block, the present invention encodes and sorts quantization method to optimize clustering performance of the subimage block after dct transform using DPCM.
2nd, adaptive smooth area quantifies, the flow chart of combining adaptive Novel Fast Fractal Image Compression Approach
The quantization table of correspondingly-sized is needed during the quantification treatment adopted to bulk smooth area.It is the table based on normal brightness quantization table and standard colorimetric quantization table to this solution, quantization table is that underlying table is formed through neighbor interpolation.For large-sized quantization table, wherein the coefficient in a certain ranks should meet first in underlying table in correspondence ranks variation relation between coefficient, and the quantization step of medium-high frequency is suitably increased on this basis.
The present invention proposes a kind of large scale and quantifies table generating method, first pass around spline interpolation and obtain in ranks the large scale quantization table that relation between coefficient meets in underlying table variation relation between coefficient in correspondence ranks, the quantization step of medium-high frequency part is then suitably increased on this basis.After the process of self adaptation piecemeal, the size of subimage block is bigger, and it is smoothed all the more, and the medium-high frequency coefficient in composing to it increases quantization degree can ensured to recover to increase substantially compression ratio on the basis of picture quality.Therefore it is appropriate to the medium-high frequency quantization step in gained large scale quantization table to tune up, increase quantization degree.
3rd, the non-smooth area of self adaptation quantifies, the flow chart of combining adaptive Novel Fast Fractal Image Compression Approach
Marginal zone and texture area are referred to as non-smooth area.The space characteristics difference of non-smooth area is huge, and the distribution of significant coefficient is also without obvious rule in corresponding spectrum.Except minority determines the important low-frequency component integrated distribution of non-smooth area overall intensity in addition to the upper left corner of spectrum, the medium, high frequency composition for characterizing its complex space feature is distributed without evident regularity in the space in addition to the upper left corner.For non-smooth area, although in its spectrum important coefficient without evident regularity in spatial distribution, from terms of amplitude angle, it is larger that these significant coefficients all show as amplitude.Adaptive quantizing method to the quantification treatment of non-smooth area is quantified according to the importance of coefficient in spectrum.
Process to DC component, what DC component was characterized is the average gray of subimage block in smooth area and non-smooth area, smooth area and non-smooth area have seriality in spatial distribution, the DC component for spatially closing on subimage block has very strong dependency, identical quantization step is adopted to DC component in smooth area and non-smooth area, the non-smooth area obtained after the process of self adaptation piecemeal is when quantifying, and the direct current quantization step that the quantization step of DC component is selected with the normal brightness recommended quantifies table in non-smooth area is identical.
Process to AC compounent, to phenogram in non-smooth area as the AC compounent of block space feature quantifies according to importance, importance is embodied adaptive quantizing by the amplitude of coefficient, and the big coefficient importance of amplitude is big, using little quantization step.The little importance of amplitude is little, using big quantization step.
During the non-smooth area of self adaptation is quantified to be used in subimage block generation adaptive block, maximum advantage is namely based on the cluster residual error that the quantization of non-smooth area can be reduced effectively in subimage block, for processing the AC compounent isolated from original image.Also there is great meaning to segmentation threshold block later and mean cluster operation.The present invention is after non-smooth area subimage block carries out dimension conversion, descending sequence to be carried out according to amplitude to all coefficients, before the big significant coefficient of amplitude comes, behind the little insignificant coefficient of amplitude comes.Sequence produces the data of each coefficient of part of records position in original spectrum simultaneously, and this partial data is referred to as address date.After decoding end inverse quantization, the coefficient after inverse quantization is put back on its position in original spectrum using address date.
4th, fast typing, the flow chart of combining adaptive Novel Fast Fractal Image Compression Approach
Setting up one based on fractals can provide variously-shaped shape library.Shape library is in itself not deposit fractal pattern, but iterated function system (IFS) coded combination of storage.Corresponding fractal graph or space structure can be reconstructed by these coded combinations.In fractal compression, constitute IFS key component be one or one group of contractive affine transform, image and the original iterative image that these contractive affine transforms are obtained under the iterated conditional of the limit it does not matter whether, and only with the relating to parameters of affine transformation.The process of fractal compression image is exactly to find out the process of such one group of affine transformation parameter.The essence of Fractal Image Compression be in image local-local between, very big affine redundancy is there is between local-entirety.Such as structural redundancy is exactly the wherein very big a kind of affine redundancy of accounting, i.e., can greatly pass through the affine of ego structure in image, or through a certain degree of conversion, so as to realize the expression of information.Fractal image coding reaches compression of images by the self similarity redundancy of each several part between this removal image inside.Various fractal structures in image are found out, and is expressed to realize the compression process of image by affine transformation.The method of this storage or transmission IFS parameters can obtain very high compression ratio.
Above-mentioned combination accompanying drawing is described in detail to embodiments of the invention, it should be appreciated that above-mentioned simply exemplary, therefore, protection scope of the present invention should be determined by the content of appending claims.
Claims (3)
1. a kind of self adaptation Novel Fast Fractal Image Compression Approach, the method is related to computer information technology field, more particularly to compression of images field, it is characterized in that：The method realizes that step is as follows:
Step 1：Each label frequency of occurrences in sub-block is counted according to original image labelling matrix, obtain maximum, maximum is compared with decision threshold, more than decision threshold, then subimage block is judged to smooth area, otherwise continue to judge whether subimage block size is standard size, it is to divide the image into multiple subimage blocks according to quaternary tree mode, subimage block is scanned one by one, obtains the current subimage block after scanning；
Step 2：Current subimage block is processed using jpeg algorithm compression, after the dimension dct transform of current subimage block 2, DC components carry out DPCM coded treatments, to sorting from big to small according to amplitude after AC component swepts, then quantify, encode, adaptive block is generated, while to the closely related coding of address date, realizing image adaptive process；
Step 3：Adaptive block according to generating sets up domain blocks storehouse, the step-size in search of domain blocks is r, and remove flat domain blocks, i.e. gray standard deviation is less than the block of certain given threshold value, the completely available gray value of flat block replaces equal to the constant block of range block average, this reduces clusters number and search space；
Step 4：According to the domain blocks storehouse segmentation range block set up, gray standard deviation is considered as into flat block less than the range block of certain threshold value n, directly it is equal to the constant block of range block average with gray value to approach, extract two features of average and variance of range block, K mean cluster is carried out using the two features, K range block class is obtained；
Step 5：Extract two features of average and variance of adaptive block, adaptive block is clustered with the cluster centre of K range block class, obtain corresponding K domain blocks class, finally K range block class and corresponding K domain blocks class are matched, fast typing compression image of the adapting to image based on K mean cluster is obtained.
2. a kind of self adaptation Novel Fast Fractal Image Compression Approach according to claim 1, is characterized in that：The present invention proposes a kind of large scale and quantifies table generating method：First pass around spline interpolation and obtain in ranks the large scale quantization table that relation between coefficient meets in underlying table variation relation between coefficient in correspondence ranks, then the quantization step of medium-high frequency part is suitably increased on this basis, after the process of self adaptation piecemeal, the size of subimage block is bigger, it is smoothed all the more, medium-high frequency coefficient in composing to it increases quantization degree can ensured to recover to increase substantially compression ratio on the basis of picture quality, therefore it is appropriate to the medium-high frequency quantization step in gained large scale quantization table to tune up, increase quantization degree.
3. a kind of self adaptation Novel Fast Fractal Image Compression Approach according to claim 1, is characterized in that：The present invention is after non-smooth area subimage block carries out dimension conversion, descending sequence is carried out according to amplitude to all coefficients, before the big significant coefficient of amplitude comes, behind the little insignificant coefficient of amplitude comes, sequence produces the data of each coefficient of part of records position in original spectrum simultaneously, this partial data is referred to as address date, after decoding end inverse quantization, the coefficient after inverse quantization is put back on its position in original spectrum using address date.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610079621.2A CN106612439A (en) | 2016-02-04 | 2016-02-04 | Adaptive fast fractal image compression method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610079621.2A CN106612439A (en) | 2016-02-04 | 2016-02-04 | Adaptive fast fractal image compression method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106612439A true CN106612439A (en) | 2017-05-03 |
Family
ID=58615311
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610079621.2A Pending CN106612439A (en) | 2016-02-04 | 2016-02-04 | Adaptive fast fractal image compression method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106612439A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108769704A (en) * | 2018-06-04 | 2018-11-06 | 浙江工业大学 | A kind of discrete cosine transform method for compressing image based on dynamic power analysis |
| CN109391818A (en) * | 2018-11-30 | 2019-02-26 | 昆明理工大学 | A kind of fast search Fractal Image Compression Approach based on dct transform |
| CN109493345A (en) * | 2018-10-22 | 2019-03-19 | 太原科技大学 | A kind of quick Processing Algorithm of computer digital image |
| CN109982095A (en) * | 2019-03-20 | 2019-07-05 | 南宁师范大学 | Fractal Image Compression Coding method based on CNN and GEP |
| CN112184732A (en) * | 2020-09-27 | 2021-01-05 | 山东炎黄工业设计有限公司 | Intelligent image processing method |
| TWI770441B (en) * | 2018-12-17 | 2022-07-11 | 日商佳能股份有限公司 | Image encoding device, image decoding device, control method and program thereof |
| WO2022247735A1 (en) * | 2021-05-28 | 2022-12-01 | 于江鸿 | Data processing method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547261A (en) * | 2010-12-24 | 2012-07-04 | 上海电机学院 | Fractal image encoding method |
| CN104780368A (en) * | 2015-04-28 | 2015-07-15 | 华东交通大学 | Self-adaptation sampling method based on image statistical information |
-
2016
- 2016-02-04 CN CN201610079621.2A patent/CN106612439A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547261A (en) * | 2010-12-24 | 2012-07-04 | 上海电机学院 | Fractal image encoding method |
| CN104780368A (en) * | 2015-04-28 | 2015-07-15 | 华东交通大学 | Self-adaptation sampling method based on image statistical information |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108769704A (en) * | 2018-06-04 | 2018-11-06 | 浙江工业大学 | A kind of discrete cosine transform method for compressing image based on dynamic power analysis |
| CN109493345A (en) * | 2018-10-22 | 2019-03-19 | 太原科技大学 | A kind of quick Processing Algorithm of computer digital image |
| CN109391818A (en) * | 2018-11-30 | 2019-02-26 | 昆明理工大学 | A kind of fast search Fractal Image Compression Approach based on dct transform |
| CN109391818B (en) * | 2018-11-30 | 2021-12-10 | 昆明理工大学 | DCT (discrete cosine transformation) -based fractal image compression method for rapid search |
| TWI770441B (en) * | 2018-12-17 | 2022-07-11 | 日商佳能股份有限公司 | Image encoding device, image decoding device, control method and program thereof |
| TWI788268B (en) * | 2018-12-17 | 2022-12-21 | 日商佳能股份有限公司 | Image encoding device, image decoding device, control method and program thereof |
| CN109982095A (en) * | 2019-03-20 | 2019-07-05 | 南宁师范大学 | Fractal Image Compression Coding method based on CNN and GEP |
| CN112184732A (en) * | 2020-09-27 | 2021-01-05 | 山东炎黄工业设计有限公司 | Intelligent image processing method |
| CN112184732B (en) * | 2020-09-27 | 2022-05-24 | 佛山市三力智能设备科技有限公司 | Intelligent image processing method |
| WO2022247735A1 (en) * | 2021-05-28 | 2022-12-01 | 于江鸿 | Data processing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106612439A (en) | Adaptive fast fractal image compression method | |
| CN102550027B (en) | Locally variable quantization and hybrid variable length coding for image and video compression | |
| Zhou et al. | DCT-based color image compression algorithm using an efficient lossless encoder | |
| Zhang et al. | Rate-distortion optimized sparse coding with ordered dictionary for image set compression | |
| Liu et al. | Prior-based quantization bin matching for cloud storage of JPEG images | |
| Moinuddin et al. | Efficient algorithm for very low bit rate embedded image coding | |
| Chu et al. | A digital image watermarking method based on labeled bisecting clustering algorithm | |
| CN102307303B (en) | Ternary-representation-based image predictive coding method | |
| Mulla et al. | Comparison of Different Image Compression Techniques | |
| Khmelevskiy et al. | Model of Transformation of the Alphabet of the Encoded Data as a Tool to Provide the Necessary Level of Video Image Qualityi in Aeromonitoring Systems. | |
| Huang et al. | An improved LBG algorithm for image vector quantization | |
| CN106331719A (en) | K-L transformation error space dividing based image data compression method | |
| Begum et al. | An efficient algorithm for codebook design in transform vector quantization | |
| Poolakkachalil et al. | Comparative analysis of lossless compression techniques in efficient DCT-based image compression system based on Laplacian Transparent Composite Model and An Innovative Lossless Compression Method for Discrete-Color Images | |
| CN102316324B (en) | Image coding prediction method based on local minimum entropy | |
| CN101064844A (en) | Method for performing matching compression to image using rotary compressed codebook | |
| Chaker et al. | An improved image retrieval algorithm for JPEG 2000 compressed images | |
| CN108259896B (en) | Columbus-Rice initial parameter self-adaptive decision method utilizing coefficient distribution characteristics | |
| Jha et al. | DEMD-based video coding for textured videos in an H. 264/MPEG framework | |
| Mahdi et al. | Image Compression using Polynomial Coding Techniques: A review | |
| Feng et al. | Image coding based on classified vector quantisation using edge orientation patterns | |
| Shoa et al. | Optimized atom position and coefficient coding for matching pursuit-based image compression | |
| Gadha et al. | An image compression and classification method by reducing the blocking effect | |
| Feng et al. | An efficient hybrid feature for edge-preserving based on block truncation coding and tree-structured vector quantization with edge orientation classification of bit-maps | |
| Guo et al. | Content-based image retrieval with ordered dither block truncation coding features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170503 |